I think another possibility is that you are getting the output of a screen
from the manufacturer. Making the capacitors then pulling those that are 1%,
5%, and 10% from the mix. This leaves the ends of the screen available to
those purchasing the 20% parts. 

It has been years, but I did experience this from a maker of film caps.
Sometimes we would get all parts above the nominal other times all below the
nominal. Never, nominal. When we paid for 5%, we got parts around nominal. 

-----Original Message-----
From: TechNet [mailto:[log in to unmask]] On Behalf Of Ioan Tempea
Sent: Tuesday, July 01, 2008 11:18 AM
To: [log in to unmask]
Subject: Re: [TN] Ceramic Cap Value

Hi Ken,

We've seen this phenomenon twice in the past 4 months, high rate of ICT
failure on ceramic caps and had the same issue, replacing the parts
sometimes worked and sometimes didn't. We even have a painful open case on
this matter.

I need more clarifications from the group regarding this issue and how to
get around it.

The paper on aging asks to take the phenomenon into account when designing
the circuitry.
Does this mean the high failure rate Ken and I witnessed is borderline
design?

Can it be marginal testing method also?

Can it be anything else that design or testing?

Thanks,

Ioan


-----Original Message-----
From: TechNet [mailto:[log in to unmask]] On Behalf Of Bruce Tostevin
Sent: Monday, June 30, 2008 9:37 AM
To: [log in to unmask]
Subject: Re: [TN] Ceramic Cap Value

Ken,

As John Maxwell pointed out, it sounds like you may be seeing the normal
effects of capacitor dielectric aging.  It's a common stumbling block with
parts at incoming inspection, test, etc.  There's a pretty good one page
summary with a handy formula at:

http://www.atceramics.com/technicalnotes/circuit_designer.asp

Click on Dielectric Aging Phenonema.  The aging aspects of ceramic
capacitors are also covered in EIA-521.

A quick way to see if this is what's going on with your parts is to preheat
then a little bit and float a few parts in a solderpot for 5-10 seconds and
remeasure them after they cool. That'll usually bring the value up enough to
confirm the issue.

Bruce Tostevin
Benchmark Electronics
Hudson, New Hampshire

-----Original Message-----
From: TechNet [mailto:[log in to unmask]] On Behalf Of Ken Bloomquist
Sent: Friday, June 27, 2008 5:14 PM
To: [log in to unmask]
Subject: [TN] Ceramic Cap Value

Happy Friday,

I need some help understanding how you test for the value of ceramic
capacitors. We build a board that has some 0603, .470 uF ceramic caps,
+/-
10%. The assembly that these parts are on has been having a higher failure
rate at our ICT then we've seen in the past. We tracked the failure to this
cap which had a low reading. We replaced the cap and the unit passed ICT
just fine.

We measured the cap using our HP 4262A LCR Meter and sure enough it was
reading low, around .398 uF.

We pulled some new parts off the reel and they read low also ranging
anywhere from .398 to .425. We suspected that we had a bad reel of parts
even thought they were from a respected cap manufacturer.

Being curious we tested some more reels from different lot codes and found
the same thing. Being even more curious we tested some other parts from the
same manufacturer and found them to be low by around 5% - 7% and in some
cases slightly below 10%. We then tested some other parts from a different
manufacturer and they tested low also.

Now we are suspecting our test equipment so we went to an independent lab
and had them tested on an Andeen Hagerling 2500A Bridge. Guess what, we got
the same low readings.

After all that my question is, what are we doing wrong to get such low
readings on these ceramic caps? It just doesn't seem likely that four
different parts from three different major suppliers tested on two different
pieces of equipment would all be that far down on the low end.

Any education in capacitor testing and reading would surely be appreciated.

Best regards,

KennyB

---------------------------------------------------
Technet Mail List provided as a service by IPC using LISTSERV 15.0
To unsubscribe, send a message to [log in to unmask] with following text in
the BODY (NOT the subject field): SIGNOFF Technet
To temporarily halt or (re-start) delivery of Technet send e-mail to [log in to unmask]: SET Technet NOMAIL or (MAIL)
To receive ONE mailing per day of all the posts: send e-mail to [log in to unmask]: SET Technet Digest
Search the archives of previous posts at: http://listserv.ipc.org/archives
Please visit IPC web site http://www.ipc.org/contentpage.asp?Pageid=4.3.16 for additional information, or contact Keach Sasamori at [log in to unmask] or 847-615-7100 ext.2815
-----------------------------------------------------